Forecasting accuracy influence on logistics clusters activities: The case of the food industry.

The current approaches to supply chain management generate large amounts of food waste due to the growing urbanization levels, increasing consumer demand for organic products and the growth of e-commerce distribution channel. These trends require the organizations to rethink their approaches to supply chain management, so that they could cope with the upcoming challenges. This research paper focuses on the ways a logistics cluster can provide the abilities to share information and thus improve the forecasting accuracy. The main novelty of the paper has been to work out a collaborative technological strategy which promotes information sharing in order to improve forecasting accuracy and inventory control for better alignment of demand and supply. The secondary contribution of the article is the determination of the influence of information sharing on forecasting accuracy in different market sizes, types and the consideration of consumer integration. Lastly, a sensitivity analysis has been completed to identify the optimal size of the logistics cluster the determination of which provides support when implementing the proposed strategy in practice. For the results validation we have implemented an agent-based model of the food supply chain. Our results have confirmed that information sharing increases the forecasting accuracy in multiple scenarios. Moreover, consumer integration is beneficial in a perfect competition market; however, its positive effect is less significant in an oligopoly market. These findings should be taken into consideration when developing e-commerce business strategy and forming logistics clusters. The usage of machine learning algorithms in the forecasting process provides adaptation capabilities for the supply chain members. The adaptation emerges as system resilience and it improves the alignment of demand and supply, reduces food waste levels and maintains higher nutrition value. The proposed strategy can ensure a long-term sustainable development to logistics cluster members. • The majority of supply chain management approaches have been developed in astable business environment. • Supply chain collaboration tends to fail due to lack of collaborative technologies and strategies. • The inability to align demand and supply in the food supply chain generates a lot of food waste and reduces nutrition value. • An agent-based model identified how collaborative demand forecasting improves alignment of demand and supply. • Application of machine learning provides adaptation abilities for supply chain members, and ensures system resilience. [ABSTRACT FROM AUTHOR]